WO2020188748A1

WO2020188748A1 - Surveillance system, information processing device, fall detection method, and non-temporary computer readable medium

Info

Publication number: WO2020188748A1
Application number: PCT/JP2019/011469
Authority: WO
Inventors: 伸一宮本
Original assignee: 日本電気株式会社
Priority date: 2019-03-19
Filing date: 2019-03-19
Publication date: 2020-09-24
Also published as: JPWO2020188748A1; US20220163671A1; JP7231011B2

Abstract

An objective of the present invention is to provide a surveillance system, an information processing device, a fall detection system, and a non-temporary computer readable medium capable of improving the precision of detecting a person who has fallen inside a railroad crossing. The surveillance system according to the present disclosure comprises: a LiDAR sensor (10) which irradiates a surveillance region (60) with a plurality of laser beams having different ranges, and outputs a detection signal indicating a detection status of an object (40) by the respective laser beams; and an information processing device (20) which determines that the object (40) has fallen when the detection status for the lasers having a range shorter than a prescribed range indicates that the object (40) has been detected and the number of laser beams detecting the object (40) has decreased.

Description

Surveillance systems, information processing equipment, inversion detection methods, and non-temporary computer-readable media

This disclosure relates to monitoring systems, information processing devices, fall detection methods, and non-temporary computer-readable media.

In order to reduce accidents inside railroad crossings, monitoring devices that monitor railroad crossings are used. The monitoring device detects a person existing in the railroad crossing by using information obtained from a camera, radar, or the like inside the railroad crossing.

Patent Document 1 discloses a configuration of a railroad crossing obstacle detection device that can reliably detect an object whose height, area, size, etc., which should be detected as an obstacle, is changed as an obstacle in the railroad crossing. ing. The railroad crossing obstacle detection device of Patent Document 1 emits a transmitted wave from a millimeter wave radar sensor or the like toward an obstacle detection region, and detects the distance to the obstacle, the size of the obstacle, etc. using the reflected wave. To do. Further, the obstacle detection device defines the emission range of the transmitted wave of the millimeter wave radar so as to cover the obstacle detection area in the railroad crossing from the center of the antenna portion of the millimeter wave radar sensor.

Japanese Unexamined Patent Publication No. 2016-155482

The transmitted wave emitted from the railroad crossing obstacle detection device disclosed in Patent Document 1 needs to be emitted substantially horizontally with the ground in order to cover the obstacle detection area. By emitting the transmitted wave substantially horizontally with the ground, the transmitted wave reaches the area at the end of the obstacle detection area. However, when the transmitted wave is emitted substantially horizontally with the ground, there is a problem that the transmitted wave does not hit the person who has fallen in the railroad crossing and the person who has fallen cannot be detected.

An object of the present disclosure is to provide a monitoring system, an information processing device, a fall detection method, and a non-temporary computer-readable medium capable of improving the accuracy of detecting a person who has fallen in a railroad crossing.

The monitoring system according to the first aspect of the present disclosure includes a LiDAR sensor that irradiates a plurality of laser beams having different reach to a monitoring region and outputs a detection signal indicating an object detection status in each laser beam, and a predetermined LiDAR sensor. Information processing that determines that the object has fallen when the detection status of the laser beam having a reach shorter than the reach distance indicates that the object has been detected and the number of laser beams that have detected the object decreases. It is equipped with a device.

In the information processing apparatus according to the second aspect of the present disclosure, a plurality of laser beams having different reach are emitted from the LiDAR sensor to the monitoring region, and a detection signal indicating the detection status of an object in each laser beam is transmitted from the LiDAR sensor. When the detection status of the receiving communication unit and the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected, and the number of laser beams that have detected the object decreases, the object It is provided with a determination unit for determining that the laser has fallen.

In the fall detection method according to the third aspect of the present disclosure, a plurality of laser beams having different reach distances are irradiated from the LiDAR sensor to the monitoring region, and a detection signal indicating the detection status of an object in each laser beam is transmitted from the LiDAR sensor. The object has fallen when it is received and the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. Is determined.

In the program according to the fourth aspect of the present disclosure, a plurality of laser beams having different reach are emitted from the LiDAR sensor to the monitoring area, and a detection signal indicating the detection status of an object in each laser beam is received from the LiDAR sensor. , It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. Let the computer do what you want to do.

According to the present disclosure, it is possible to provide a monitoring system, an information processing device, a fall detection method, and a non-temporary computer-readable medium that can improve the accuracy of detecting a person who has fallen in a railroad crossing.

It is a block diagram of the monitoring system which concerns on Embodiment 1. FIG. It is a block diagram of the monitoring system which concerns on Embodiment 2. FIG. It is a block diagram of the LiDAR sensor which concerns on Embodiment 2. FIG. It is a block diagram of the information processing apparatus which concerns on Embodiment 2. FIG. It is a figure explaining the determination process of whether or not there is a person who has fallen down which concerns on Embodiment 2. FIG. It is a figure which shows the flow of the detection process of the fallen person which concerns on Embodiment 2. FIG. It is a figure which shows the flow of the detection process of the fallen person which concerns on Embodiment 3. FIG. It is a block diagram of the information processing apparatus which concerns on each embodiment.

(Embodiment 1)
Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. A configuration example of the monitoring system according to the first embodiment will be described with reference to FIG. The monitoring system of FIG. 1 is mainly used for monitoring the state inside a railroad crossing. Specifically, the monitoring system of FIG. 1 may be used to detect a fallen person at a railroad crossing within the monitoring area 60.

The monitoring system has a LiDAR (Light Detection and Ringing) sensor 10 and an information processing device 20. The LiDAR sensor irradiates the monitoring area 60 with a plurality of laser beams having different reach, and outputs a detection signal indicating the detection status of the object 40 in each laser beam. Regarding the object 40, FIG. 1 shows a state in which the object 40 is walking and a state in which the object 40 is overturned. It can be said that the state in which the object 40 is overturned is lower than the state in which the object 40 is walking. The state in which the object 40 is walking is shown by a solid line, and the state in which the object 40 is overturned is shown by a dotted line.

The monitoring area 60 is a space on the ground surface 50, and may be, for example, a space including a railroad crossing. Further, although it is shown in FIG. 1 that the upper end of the monitoring area 60 is defined, the upper end may not be defined.

A plurality of laser beams having different reach distances may be paraphrased as each laser beam having a different irradiation angle from other laser beams. The irradiation angle may be determined by a plane parallel to the ground surface 50 and the irradiation direction of the laser beam. Alternatively, the irradiation angle may be determined by a plane orthogonal to the ground surface 50 and the irradiation direction of the laser beam. The reach may be the distance between the LiDAR sensor 10 and the reach point of the laser beam on the ground surface 50. Alternatively, the reachable distance may be the distance between the contact point between the LiDAR sensor 10 and the ground surface 50 and the reachable point of the laser beam on the ground surface 50.

The laser light emitted from the LiDAR sensor 10 scatters when it hits the object 40 and becomes reflected light. The reflected light is used for determining the presence or absence of the object 40 and for measuring the distance to the object 40. When the LiDAR sensor 10 receives the reflected light, it outputs a detection signal indicating the detection status of the object 40 to the information processing device 20. The detection status is, for example, a status indicating whether or not the object 40 has been detected. However, even if the LiDAR sensor 10 receives the reflected light, it cannot determine whether the recognized object 40 is walking or falling, and only recognizes that some object 40 has been detected. ..

The LiDAR sensor 10 may collectively set the detection status of the object 40 in each irradiated laser beam into one signal, or may set the detection status in a different signal for each laser beam. In the detection signal, information indicating whether or not the object 40 is detected may be set, or information indicating whether or not reflected light is received may be set for each laser beam.

The LiDAR sensor 10 transmits a detection signal to the information processing device 20 via the network 30. The network 30 may be, for example, an IP network. Specifically, the network 30 may be the so-called Internet or an intranet which is a closed network in a company or the like. Alternatively, the LiDAR sensor 10 and the information processing device 20 may directly communicate with each other using a wired cable, short-range wireless communication, or the like. The short-range wireless communication may be, for example, a wireless LAN (Local Area Network) or Bluetooth (registered trademark).

The information processing device 20 may be a computer device that operates by the processor executing a program stored in the memory. The information processing device 20 may be, for example, a server device.

The information processing device 20 receives a detection signal from the LiDAR sensor 10. The information processing apparatus 20 indicates that the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object 40 has been detected, and when the number of laser beams that have detected the object 40 decreases, the object becomes Determined to have fallen.

For example, as shown in FIG. 1, it is assumed that the object 40 is detected by three laser beams out of the four laser beams emitted from the LiDAR sensor 10. After that, when the object 40 falls down, the object 40 lies at a position close to the ground surface 50. In this case, among the plurality of laser beams emitted from the LiDAR sensor 10, the object 40 cannot be detected by the three laser beams having a long reach, and only the laser beam having the shortest reach is the object 40. Can be detected. In this way, when the laser beam that has detected the object 40 decreases and the laser beam that has a short reach detects the object 40, the LiDAR sensor 10 determines that the object 40 has fallen.

In FIG. 1, it is shown that only the laser beam having the shortest reach among the four laser beams detected the overturned object 40. For example, two of the four laser beams having the shortest reach. The laser beam may detect an overturned object. When the laser light that has detected the object 40 decreases, it may be predetermined which laser light the object 40 is determined to have fallen when the object 40 is detected.

As described above, the monitoring system of FIG. 1 can use the detection results of objects in a plurality of laser beams having different reach areas when determining whether or not the object 40 has fallen. Multiple lasers with different reach have different angles with respect to the surface of the earth. Therefore, it is possible to improve the possibility that any of the laser beams hits the overturned object as compared with the case where the laser beam is irradiated parallel to the ground surface. As a result, the information processing apparatus 20 can determine whether or not the object 40 has fallen by using the detection results of the object 40 in the plurality of laser beams.

(Embodiment 2)
Subsequently, a configuration example of the monitoring system according to the second embodiment will be described with reference to FIG. The monitoring system of FIG. 2 includes a LiDAR sensor 10, an information processing device 20, and a camera 70. The LiDAR sensor 10 and the information processing device 20 are the same as the LiDAR sensor 10 and the information processing device 20 described with reference to FIG. The functions and operations of the LiDAR sensor 10 and the information processing device 20 similar to those in FIG. 1 will not be described in detail.

The camera 70 is used to photograph the monitoring area 60. The camera 70 may be, for example, a far-infrared camera, a general digital camera, a video camera, or the like.

The camera 70 transmits the captured image generated by photographing the monitoring area 60 to the information processing device 20. The camera 70 may transmit the captured image to the information processing device 20 via the network 30. Alternatively, the camera 70 may transmit the captured image to the information processing device 20 by using a wired cable, short-range wireless communication, or the like.

Subsequently, a configuration example of the LiDAR sensor 10 according to the second embodiment will be described with reference to FIG. The LiDAR sensor 10 has an irradiation unit 11 and a communication unit 12. The irradiation unit 11 may irradiate the laser beam and receive the reflected light. The irradiation unit 11 may be composed of, for example, an amplifier, an ADC (Analog Digital Converter), a photodiode, or the like. The irradiation unit 11 may be paraphrased as being composed of, for example, a transmitter that irradiates laser light and a receiver that receives reflected light. The irradiation unit 11 irradiates a plurality of laser beams having different irradiation angles. The irradiation unit 11 may irradiate a plurality of laser beams by changing the irradiation direction of one transmitter, or may irradiate a plurality of laser beams using a plurality of transmitters having a fixed irradiation direction. Good.

The irradiation unit 11 outputs to the communication unit 12 information indicating which of the irradiated laser light the reflected light is received. The irradiation unit 11 may determine, for example, that the object 40 exists in the irradiation direction of the laser light that could receive the reflected light. For example, among the laser beams emitted from the LiDAR sensor 10 in FIG. 1, the laser beams of the first layer, the second layer, the third layer, and the fourth layer are used in order from the laser beam having the longest reach. In FIG. 1, the irradiation unit 11 first outputs information indicating that the object 40 is detected in the laser light from the first layer to the third layer to the communication unit 12. After the object 40 has fallen, the irradiation unit 11 outputs information indicating that the object 40 has been detected in the laser light of the fourth layer to the communication unit 12.

The communication unit 12 communicates with the information processing device 20 via the network 30. The communication unit 12 may be, for example, a network interface that transmits and receives data to and from the network 30. The communication unit 12 transmits a detection signal including information indicating the detection result of the object 40 to the information processing device 20. The information indicating the detection result of the object 40 may be, for example, information indicating in which layer of the laser light the object 40 was detected. The communication unit 12 may transmit a detection signal to the information processing device 20 each time it receives information from the irradiation unit 11. Alternatively, the communication unit 12 may transmit the detection signal to the information processing device 20 at a predetermined timing.

Subsequently, a configuration example of the information processing apparatus 20 according to the second embodiment will be described with reference to FIG. The information processing device 20 has a communication unit 21 and a determination unit 22. The communication unit 21 and the determination unit 22 may be software or modules whose processing is executed by the processor executing a program stored in the memory. Alternatively, the communication unit 21 and the determination unit 22 may be hardware such as a circuit or a chip.

The communication unit 21 receives the captured image from the camera 70 and receives the detection signal from the LiDAR sensor 10 via the network 30. The communication unit 21 outputs the received captured image and the detection signal to the determination unit 22.

The determination unit 22 determines whether or not there is a person who has fallen in the monitoring area 60. Here, the determination process in the determination unit 22 will be described with reference to FIG. The determination unit 22 performs detection processing of the object 40 from the captured image. For example, the determination unit 22 detects a pedestrian included in a captured image by using a learning model in which the characteristics of a pedestrian are learned by using deep learning or the like. Each time the determination unit 22 receives a photographed image from the camera 70, the determination unit 22 performs a pedestrian detection process using each photographed image as input data of the learning model. The case where the determination unit 22 can detect a pedestrian is a case where the characteristics of the pedestrian are shown in the captured image. The case where the determination unit 22 cannot detect the pedestrian may be, for example, the case where the pedestrian moves out of the monitoring area 60. Alternatively, the case where the determination unit 22 cannot detect the pedestrian may be the case where the pedestrian's characteristics are not shown in the captured image due to the pedestrian's fall. Alternatively, the case where the determination unit 22 cannot detect the pedestrian may be the case where the pedestrian cannot be distinguished from the ground surface 50 in the photographed image taken by the far-infrared camera. Specifically, when a pedestrian falls and comes into contact with the ground surface 50, the temperature difference between the pedestrian and the ground surface 50 may become small. In such a case, the pedestrian may not be distinguishable from the ground surface 50 in the image taken by the far-infrared camera.

Further, the determination unit 22 determines which layer of laser light the object 40 is detected in the detection signal received from the LiDAR sensor 10. Further, the determination unit 22 determines whether the object 40 is detected by the plurality of laser beams.

For example, the determination unit 22 determines that the pedestrian was detected in the captured image at the timing before the time t, but the pedestrian could not be detected in the captured image at the timing after the time t. .. Further, the determination unit 22 has detected the object 40 in the plurality of layers at the timing before the time t, but only the fourth layer has detected the object 40 at the timing after the time t. judge.

That is, the determination unit 22 cannot detect the pedestrian from the captured image at the time t, and further, the object 40 detected in the plurality of layers at the time t and the object 40 only in the fourth layer. It will be detected. In this way, when the object 40 detected in the plurality of layers is detected only in the fourth layer at the timing when the pedestrian cannot be detected from the captured image, the determination unit 22 determines that there is a person who has fallen. The determination unit 22 has a timing at which the pedestrian cannot be detected from the captured image and a timing at which the object 40 detected in the plurality of layers is detected in the fourth layer only. If they are substantially the same, it may be determined that some people have fallen. The cases of substantially the same are the timing when the pedestrian cannot be detected from the captured image and the timing when the object 40 detected in the plurality of layers is detected only in the fourth layer. Includes cases where is within a predetermined range.

Further, the determination unit 22 determines that the pedestrian and the object 40 are the same person when the position of the pedestrian detected in the captured image and the position of the object 40 detected by the LiDAR sensor 10 match. You may. Alternatively, if the distance between the position of the pedestrian detected in the captured image and the position of the object 40 detected by the LiDAR sensor 10 is shorter than the predetermined distance, the determination unit 22 determines the pedestrian and the object 40. May be determined to be the same person. The position of the pedestrian detected in the captured image may be specified, for example, by converting the image coordinates or the camera coordinates to the world coordinates. Image coordinates or camera coordinates may be indicated using, for example, pixel positions. World coordinates may be indicated using, for example, latitude and longitude. The determination unit 22 may hold, for example, table information indicating the correspondence between the image coordinates or the camera coordinates and the world coordinates in advance.

As for the position of the object 40 detected by the LiDAR sensor 10, for example, the detection position when the laser light of each layer detects the object 40 may be predetermined. Alternatively, the distance from the LiDAR sensor 10 to the object 40 may be estimated as the position of the object 40 detected by the LiDAR sensor 10 according to the timing of receiving the reflected light. Further, the determination unit 22 may estimate the position of the object 40 based on the estimated distance.

Subsequently, the flow of the fallen person detection process according to the second embodiment will be described with reference to FIG. First, the determination unit 22 detects a pedestrian from the captured image (S11). The determination unit 22 detects a pedestrian from the captured image by executing machine learning such as deep learning, for example.

Next, the determination unit 22 determines that the object 40 is detected by the laser light of the first layer to the third layer in the LiDAR sensor 10 (S12). The determination unit 22 determines the laser light in which the object 40 is detected by using the detection signal received from the LiDAR sensor 10. Next, the determination unit 22 is in a state where it cannot detect a pedestrian from the captured image (S13).

Next, the determination unit 22 determines whether or not the object 40 is detected only by the laser light of the fourth layer in the LiDAR sensor 10 (S14). Alternatively, the determination unit 22 may determine whether or not the LiDAR sensor 10 has detected the object 40 only with the laser light of a predetermined layer.

When the determination unit 22 recognizes that the object 40 is detected only by the laser light of the fourth layer, it determines that the pedestrian has fallen (S15). The determination unit 22 ends the process when the object 40 is detected by the laser light other than the fourth layer, or when the object 40 is not detected by any of the laser lights.

As described above, the information processing device 20 according to the second embodiment analyzes the detection status of a pedestrian in the captured image captured by the camera 70 and the detection status of the object 40 in the LiDAR sensor 10. It is possible to determine the presence or absence of a fallen person.

The plurality of laser beams emitted by the LiDAR sensor 10 have different angles and are applied to the monitoring area 60. Therefore, the LiDAR sensor 10 can detect the object 40 in the monitoring area 60 regardless of the height of the object 40.

Further, in order to detect the movement of the pedestrian falling in the monitoring area 60, the information processing device 20 detects the change of the laser beam detecting the pedestrian by the LiDAR sensor 10 and the detection of the pedestrian in the captured image. Changes in circumstances can be used. As a result, the information processing device 20 can detect that the pedestrian has fallen.

Further, the plurality of laser beams including the first layer to the fourth layer irradiated by the LiDAR sensor 10 may have a reaching point on the ground surface 50 included in the monitoring area 60. By limiting the arrival points of the plurality of laser beams emitted from the LiDAR sensor 10 to the ground surface 50 included in the monitoring area 60, it is possible to avoid the detection of an object outside the monitoring area 60.

(Embodiment 3)
Subsequently, the pedestrian detection process according to the third embodiment will be described with reference to FIG. 7. FIG. 7 shows the process after the determination unit 22 determines YES in step S14 of FIG. The state in which the determination unit 22 determines YES in step S14 of FIG. 6 is a state in which the pedestrian cannot be detected from the captured image, and the object 40 is detected only by the laser light of the fourth layer. ..

In such a state, the determination unit 22 rotates the captured image in which the pedestrian could not be detected and tries to detect the pedestrian. When learning a pedestrian using deep learning, the determination unit 22 mainly learns the standing posture of the pedestrian. Therefore, the determination unit 22 cannot detect the pedestrian when the pedestrian has fallen.

Therefore, the determination unit 22 brings the pedestrian in the fallen state closer to the standing state by rotating the captured image. Further, the determination unit 22 attempts to detect a pedestrian by applying a learning model that learns the standing posture of the pedestrian by using the rotated captured image as an input. For example, the determination unit 22 may rotate the captured image by 90 degrees, or may rotate the captured image by 90 degrees ± α (α is an arbitrary positive value).

If the determination unit 22 cannot detect a pedestrian from the rotated captured image, the determined unit 22 may rotate the captured image again and rotate the captured image a plurality of times until the pedestrian is detected. If the pedestrian can be detected by rotating within a predetermined number of times, the determination unit 22 determines that the pedestrian has fallen (S22). If the determination unit 22 cannot detect a pedestrian by rotating within a predetermined number of times, the determination unit 22 ends the process.

As described above, the information processing device 20 of the third embodiment can detect a fallen pedestrian by using the rotated photographed image. As a result, it is possible to improve the accuracy of determining whether or not the object 40 detected by the LiDAR sensor 10 is a fallen pedestrian.

Further, the case where the determination unit 22 cannot detect the pedestrian from the captured image includes the case where the pedestrian simply walks away from the monitoring area 60. Even in such a case, if the determination unit 22 rotates the captured image to try to detect a pedestrian, the processing load of the information processing device 20 increases.

On the other hand, in the third embodiment, in the case where the pedestrian cannot be detected from the captured image, when the object 40 is detected by the laser beam of the fourth layer, the determination unit 22 rotates the captured image to pedestrian. Attempts to detect. Therefore, the number of cases in which the determination unit 22 tries to detect a pedestrian by rotating the captured image can be reduced as compared with the case where the laser light information is not used. As a result, in the third embodiment, in all cases where the pedestrian cannot be detected from the captured image, it is possible to suppress an increase in the processing load as compared with the case where the captured image is rotated to try to detect the pedestrian. it can.

FIG. 8 is a block diagram showing a configuration example of the information processing device 20. Referring to FIG. 8, the information processing apparatus 20 includes a network interface 1201, a processor 1202, and a memory 1203. The network interface 1201 is used to communicate with other network node devices that make up the communication system. Network interface 1201 may be used to perform wireless communication. For example, the network interface 1201 may be used to perform wireless LAN communication specified in the IEEE 802.11 series or mobile communication specified in 3GPP (3rd Generation Partnership Project). Alternatively, the network interface 1201 may include, for example, a network interface card (NIC) compliant with the IEEE 802.3 series.

The processor 1202 reads the software (computer program) from the memory 1203 and executes it to perform the processing of the information processing apparatus 20 described using the flowchart or the sequence in the above-described embodiment. The processor 1202 may be, for example, a microprocessor, an MPU (Micro Processing Unit), or a CPU (Central Processing Unit). Processor 1202 may include a plurality of processors.

Memory 1203 is composed of a combination of volatile memory and non-volatile memory. Memory 1203 may include storage located away from processor 1202. In this case, processor 1202 may access memory 1203 via an I / O interface (not shown).

In the example of FIG. 8, the memory 1203 is used to store the software module group. The processor 1202 can perform the processing of the information processing device 20 described in the above-described embodiment by reading these software modules from the memory 1203 and executing the software modules.

As described with reference to FIG. 8, each of the processors included in the information processing apparatus 20 and the like executes one or a plurality of programs including a group of instructions for causing the computer to perform the algorithm described with reference to the drawings.

In the above example, the program can be stored and supplied to a computer using various types of non-transitory computer readable medium. Non-temporary computer-readable media include various types of tangible storage media. Examples of non-temporary computer-readable media include magnetic recording media, magneto-optical recording media (eg, magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories. The magnetic recording medium may be, for example, a flexible disk, a magnetic tape, or a hard disk drive. The semiconductor memory may be, for example, a mask ROM, a PROM (Programmable ROM), an EPROM (Erasable PROM), a flash ROM, or a RAM (Random Access Memory). The program may also be supplied to the computer by various types of temporary computer readable media. Examples of temporary computer-readable media include electrical, optical, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or a wireless communication path.

Note that this disclosure is not limited to the above embodiment, and can be appropriately modified without departing from the spirit. Further, the present disclosure may be carried out by appropriately combining the respective embodiments.

Some or all of the above embodiments may also be described, but not limited to:
(Appendix 1)
A LiDAR sensor that irradiates a monitoring area with multiple laser beams with different reach and outputs a detection signal indicating the detection status of an object in each laser beam.
It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. A monitoring system equipped with an information processing device.
(Appendix 2)
The information processing device
The monitoring system according to Appendix 1, wherein when the detection signal of the laser light having the shortest reach among the plurality of laser lights detects the object, it is determined that the object has fallen.
(Appendix 3)
The LiDAR sensor is
The monitoring system according to Appendix 1 or 2, wherein the irradiation directions of the plurality of lasers are controlled so that the plurality of laser lights reach the ground surface in the monitoring area.
(Appendix 4)
Further equipped with a camera for capturing the surveillance area,
Information processing equipment
When the object was recognized from the photographed image taken by the camera and the object could not be recognized in the photographed image which could recognize the object, the object fell down. The monitoring system according to any one of Appendix 1 to 3, wherein the monitoring system is determined to be.
(Appendix 5)
The information processing device
When the timing at which the number of laser beams that detect the object decreases and the timing at which the object cannot be recognized in the captured image that recognized the object are within a predetermined range, the said. The monitoring system according to Appendix 4, which determines that an object has fallen.
(Appendix 6)
The information processing device
The captured image that can no longer recognize the object is rotated, and the captured image that is rotated and a learning model that learns the state in which a person is standing are used to recognize the object from the captured image. The monitoring system according to Appendix 4 or 5, which determines that the object has fallen when it can be done.
(Appendix 7)
A communication unit that irradiates a monitoring area with a plurality of laser beams having different reach from the LiDAR sensor and receives a detection signal indicating an object detection status in each laser beam from the LiDAR sensor.
It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. An information processing device including a determination unit.
(Appendix 8)
The determination unit
The information processing device according to Appendix 7, wherein when the detection signal of the laser light having the shortest reach among the plurality of laser lights detects the object, it is determined that the object has fallen.
(Appendix 9)
A plurality of laser beams having different reach are emitted from the LiDAR sensor to the monitoring area, and a detection signal indicating the detection status of an object in each laser beam is received from the LiDAR sensor.
It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. , Fall detection method.
(Appendix 10)
A plurality of laser beams having different reach are emitted from the LiDAR sensor to the monitoring area, and a detection signal indicating the detection status of an object in each laser beam is received from the LiDAR sensor.
When the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases, it is determined that the object has fallen. A non-transitory computer-readable medium that contains a program that causes a computer to do things.

Note that this disclosure is not limited to the above embodiment, and can be appropriately modified without departing from the spirit.

10 LiDAR sensor 11 Irradiation unit 12 Communication unit 20 Information processing device 21 Communication unit 22 Judgment unit 30 Network 40 Object 50 Ground surface 60 Monitoring area 70 Camera

Claims

A LiDAR sensor that irradiates a monitoring area with multiple laser beams with different reach and outputs a detection signal indicating the detection status of an object in each laser beam.
It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. A monitoring system equipped with an information processing device.
The information processing device
The monitoring system according to claim 1, wherein when the detection signal of the laser light having the shortest reach among the plurality of laser lights detects the object, it is determined that the object has fallen.
The LiDAR sensor is
The monitoring system according to claim 1 or 2, wherein the irradiation directions of the plurality of lasers are controlled so that the plurality of laser lights reach the ground surface in the monitoring area.
Further equipped with a camera for capturing the surveillance area,
Information processing equipment
When the object was recognized from the photographed image taken by the camera and the object could not be recognized in the photographed image which could recognize the object, the object fell down. The monitoring system according to any one of claims 1 to 3, which determines that.
The information processing device
When the timing at which the number of laser beams that detect the object decreases and the timing at which the object cannot be recognized in the captured image that recognized the object are within a predetermined range, the said. The monitoring system according to claim 4, wherein it is determined that the object has fallen.
The information processing device
The captured image that can no longer recognize the object is rotated, and the captured image that is rotated and a learning model that learns the state in which a person is standing are used to recognize the object from the captured image. The monitoring system according to claim 4 or 5, wherein if the object can be determined to have fallen.
A communication unit that irradiates a monitoring area with a plurality of laser beams having different reach from the LiDAR sensor and receives a detection signal indicating an object detection status in each laser beam from the LiDAR sensor.
It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. An information processing device including a determination unit.
The determination unit
The information processing apparatus according to claim 7, wherein when the detection signal of the laser beam having the shortest reach among the plurality of laser beams detects the object, it is determined that the object has fallen.
A plurality of laser beams having different reach are emitted from the LiDAR sensor to the monitoring area, and a detection signal indicating the detection status of an object in each laser beam is received from the LiDAR sensor.
It is determined that the object has fallen when the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that have detected the object decreases. , Fall detection method.
A plurality of laser beams having different reach are emitted from the LiDAR sensor to the monitoring area, and a detection signal indicating the detection status of an object in each laser beam is received from the LiDAR sensor.
When the detection status of the laser beam having a reach shorter than the predetermined reach indicates that the object has been detected and the number of laser beams that detect the object decreases, it is determined that the object has fallen. A non-transitory computer-readable medium that contains a program that causes a computer to do things.